Develop a coherent hypothesis, in mathematical terms, of the pathogenesis of breast cancer that is sufficiently explicit to be rigorously tested against current data, and which can lead to testable predictions. Such a model will have heuristic value for a wide variety of workers dealing with the breast cancer problem. It will be capable of progressive modification by new data.